Policy and community health

This is a draft document. Please feel free to make edits, and understand that content may substantively change. If appropriate, discuss at Talk:Policy and community health.

This article examines how policy change would impact the community.

Quantitative data indicates that policy has had no impact on community growth, and that community barriers are created at a behavioral level. The impact of group behavior is confirmed in a notable case involving SimonP.

At the board level and community level, there are concerns that the task force would lack the authority and credibility to make policy changes on behalf of the community. Executive Director Sue Gardner advises that the task force focus on designing a representative body that could resolve outstanding policy issues.

Current Wikimedia Position

Sue Gardner, Executive Director of WMF, cautioned against recommending policy changes:

Let's imagine this group had a year or more ... to develop dozens of recommendations for changes to Wikimedia project policies, designed to support community health. If that happened, I think the most the Board would be able to do would be to recommend that all the various Wikimedia language versions / projects take a look at the recommendations, and consider adopting them. I don't think the Board would be comfortable mandating adoption.

Why? My guess is that the Board would feel ill-equipped, itself, to evaluate the recommendations. It is the body delegated by the editing community to be responsible for advancement of the mission, but it has only nine members (six from “the community”), and they do not collectively reflect very many projects or language versions. (And nor does this task force. Which is not your fault: you weren't designed to be representative.) So my guess is that the Board would feel uncomfortable mandating adoption of your recommendations.

Other volunteers have echoed that policy recommendations run the risk of not representing the community. Sue Gardner advised against mandating policy changes, and suggested the task force design a representative body that could deal with policy issues:

... rather than focusing your energy on the development of recommendations for new meta-level (cross-project, cross-language) policy changes (or maybe in addition to it)......... your group might better focus energy on developing a recommendation for a meta-level body that would have the necessary credibility and moral authority to mandate changes.

Research on policy and community health

Quantitative research

A study from Travis Kriplean shows that although policy citation has increased over time. However, the relative proportions of the different policies have been relatively stable. No particular area of policy has gained a greater share of community attention. The only exception is the "major shift in emphasis towards attribution work -- making sure that Wikipedia is a source of knowledge where facts are attributed to sources. This indicates a change in concern regarding content quality, which may also be a response to an increasing amount of negative press about reliability". (The study cites "reliable sources" as an example of an attribution policy.)

A quantitative study from José Felipe Ortega Soto examined the community growth patterns of the different language Wikipedias. He concluded that growth patterns always stabilize, regardless of policy. He elaborated on this in his interview:

Our results just showed us that this is true: policies have no influence on this stabilization effects. Instead, what is changing is the behavioral patterns in the community.

Having found that rules are not having an impact, he focused on a changing culture or "spirit" of Wikipedia:

So, (and this is a delicate issue) my point is that the problem is not in adding or removing policies. The problem is that new people coming does not share the same spirit of veteran admins (e.g. assume good faith, obviate all rules if it is in favor of improving the project, search for consensus and leave aside personal disputes, etc.). They put more difficulties to open new articles, edit the articles under "their own control", and debate with other people.

The interview ends with advice to grow the community, echoing themes such as outreach, usability, social networking tools, and mentorships.

Qualitative research

Sue Gardner pointed to an interesting news story, where a veteran editor scaled back their contributions in part because of an article being deleted:

Pulsifer is also still peeved that a piece he contributed comparing the U.S. and Canadian governments was deleted in 2007 because of concern there weren't accompanying articles comparing Canada's government with the hundreds of other nations in the world.

This story makes for an interesting case study. SimonP's experience reinforces Ortega's conclusion that community health problems are the result of Wikipedia behavioral patterns, rather than any formal set of rules.

• The article in question was highly referenced with reliable third-party academic sources. There were relatively few unverified facts.

• The article was deleted indirectly, due to a deletion discussion about an entirely different article. The deletion discussion focused on problems with original research and neutrality, despite the article's numerous references. It does not appear that SimonP was notified of the deletion discussion.

• The incident led to infighting between administrators. SimonP unilaterally exercised his administrative privileges to undelete the article. He was then warned by another administrator for misusing his administrative authority.

These events show that community problems can be the result of group behavior, rather than policies. The article met policy, but was deleted on the whims of several editors, through improper channels. This created a negative environment for one volunteer, and led to in-fighting between administrators.

Wide disagreement about policy

Here is a quick survey of the breadth of policy proposals to improve Wikipedia. There is no consensus about how to improve policy.

• Proposal:Allow original research
• Proposal:Replace NPOV with Identifiable-POV
• Proposal:Verifiability policy considered harmful
• Proposal:Reliable content
• Proposal:Intelligent Editors
• Proposal:Hire experts and don't allow changes from original
• Proposal:Drop the idea that everyone can contribute knowledge
• Proposal:Eliminate vandals on sight
• Proposal:Encouraging process improvement (discusses failures to adapt, including the deletion process)

History of Policy Changes

This section is largely anecdotal. Feel free to update it with a neutral assessment of how policy making has changed.

Early in Wikipedia's history, it was possible to write a guideline, and tag it as such to see if it would stick. Frivolous policy was held in check by concerns about bureaucracy and instruction creep, and continues to be held in check today.

As Wikipedia has matured, people have recognized the need to update policy. There have been discussions that have focused on adding new policies, simplifying confusing policies, updating policies to reflect changing standards, and trying to reign in disruptive editor behavior. However, as time has gone on, this has become more and more difficult.

Wikipedia is now larger and more diverse than ever. Guidelines have been made in the past with 20 editors who share a common goal. It is not unusual to see guidelines that are discussed by nearly one hundred editors, many of whom have different views of how Wikipedia should evolve. By default, if a consensus cannot be made for a change, then nothing changes. Even a small minority of editors can prevent a consensus from forming. Whereas the process can work if people seek a compromise, there is evidence that the culture of Wikipedia has shifted away from compromise. To prevent policy changes, it only takes a few editors who are prepared to voice their disagreement and stick to it.

The end result is that it has become nearly impossible to amend, change, or simplify policy.